The practice approach to language

Saussure writes that the difficulty in delimiting language as an object of analysis is that “[t]he object is not given in advance of the viewpoint. Far from it. Rather one might say that it is the viewpoint adopted that creates the object” (1915/1972:8). This is why a phonetic transcript looks nothing like a syntax tree, which looks nothing like an interview, which looks nothing like a nod, or a twitch, or a smile, or the rhythms of argumentation that feel like home. And yet, all of these objects cohere under different analytic perspectives. Structuralism imbues language with this kind of iridescence,Footnote ³ a different object coming into view depending on the angle of approach and the distance one assumes. The practice approach to language (PAL) is resolutely structuralist in this way. It asks its practitioners to circle around an encounter, utterance, or event, approaching from various angles and distances, with an eye to the suggested outlines of a complex, relational object. Formal, phenomenological, and sociohistorical perspectives are all likely to be relevant, though none of these perspectives on its own will yield an object of analysis adequate to the communicative event as a whole (Hanks Reference Hanks1996:230). PAL is unlike Saussurian structuralism, however, in that its wholes, or totalities, are part of what must be apprehended a posteriori rather than what is posited a priori. Key analytics established for this purpose include the semantic field, the deictic field, and the social field (See below). These constructs are defined according to the distinct principles that govern them. Each one provides a way of looking or an angle to be taken in attempts to discern the contours of a complex, relational whole.

PAL draws on Bourdieu's practice theory (Bourdieu Reference Bourdieu1990), treating communicative practices as a kind of social practice. Practice theory describes how binaries such as objectivity/subjectivity, mechanism/finalism, and structure/agency are dialectically mediated in practice. A first degree reading of Bourdieu yields a vision of the social order reflected in the image of language. Hanks opts instead for a second degree reading that treats linguistic systems as one category of objective structures, which through the dialectics of use, structure the fields within which they are instantiated (Hanks Reference Hanks2005b:72–73). This is a central part of his notion of “communicative practice” (1996), which as a framework, accounts for the seemingly contradictory fact that under a formal perspective, language appears systematic, and autonomous, and yet, when viewed from the perspective of the language user, it appears to be invaded on all sides by nonlinguistic phenomena. In this model, analysis of communicative practices is guided by a triadic heuristic that points toward mutual dependencies between structure, activity, and common-sense notions that social actors have about both (ibid.). This framework was then developed into the broader practice approach to language, which directs analysis toward the processes whereby particular social orders come to seem concrete and natural and how, apart from the intentions of individuals, those orders are replicated and transformed (Hanks Reference Hanks2005a, Reference Hanks2005b, Reference Hanks2009, Reference Hanks2010).

From the perspective of PAL, formal analysis of language holds the potential to break into circles of naturalness, illuminating the communicative processes through which naturalness is generated in particular, historically situated circumstances. For example, Hanks shows the key role translation played in the phase of Spanish Colonization in Yucatán Mexico known as “the peaceful conquest” beginning in 1547. Through careful linguistic analysis of historical texts, he argues that in the process of translation, the surface of the Maya language remained intact, while the fields to which it articulated, shifted radically beneath it. Colonial conduct, colonial organization of built space, and colonial language mutually reinforced one another, creating an internally consistent colonial world, which was eventually sealed off in its own self-evidence (Hanks Reference Hanks2010:365). However, as the circle was closing, this world was subtly being appropriated and made thoroughly Maya.

It was this Mayan, Christian God, who spoke to Maya rebels and told them to cast off the Spanish. Hanks explains:

Approaches to language that at first glance would appear to have something in common with a practice approach—such as cognitive linguistics, grammaticalization theory, and integrationist linguistics—will never lead the analyst to a project in which processes of colonization and missionization are understood in new ways through linguistic analysis of historical texts. In PAL, linguistic systems are one of several views taken on a historically specific a posteriori whole. In these alternate approaches, the world in which language exists and changes is taken into account insofar as it explains facts about language. This is a fundamental difference, which derives in part from the kinds of questions that are asked at the outset.

In contrast to cognitive approaches to language (Langacker Reference Langacker2008;Footnote ⁴ Liddell Reference Liddell and McNeill2000;Footnote ⁵ Fauconnier & Turner Reference Fauconnier and Turner2002Footnote ⁶), the practice approach asks not what universal human capacities make a given utterance possible, but how relations between embodied, historically situated actors, and the deictic and social fields they are embedded in, make a given utterance feasible, and relevant (Hanks Reference Hanks1996:231). These relations reveal themselves when the analyst looks at a phenomenon through the constructs of habitus and field (see below). These relations then form a complex object, which only comes into view under several distinct perspectives—all aimed at understanding a total event (such as religious conversion and colonization in sixteenth century Yucatán). Thus although there is overlap in cognitive and practice approaches, each demands a great amount of detail where the other makes vague suggestions. For example, both practice and cognitive approaches to language posit form-meaning correspondences, which are available to the language user prior to the interaction. However, PAL urges the analyst to enact a bidirectional procedure, whereby form-meaning correspondences are drug at length through the particularities of the fields in which they are embedded and then extracted again—transformed but still comparable to systems in other languages. Cognitive linguistics, in contrast, is unconcerned with embedded particularities and instead aims to understand the universal cognitive capacities of humans. A practice approach seeks to demonstrate how an utterance is feasible and relevant in a world of social facts, whereas a cognitive approach seeks to show the cognitive capacities that make an actually occurring utterance possible. If one instance occurs in a natural setting, this is evidence for a cognitive linguist. From the perspective of PAL, there must be density and repetition, observed over long periods of time, so that what is feasible, relevant, and expectable from the perspective of language users is intuitive for the researcher. The researcher then goes beyond this intuition in the analysis. The aim at this point is not to add social context to linguistic form, correlate social and linguistic categories, or describe linguistic aspects of social activities, but to find a way in to a historically particular moment in a world by breaking the seals of naturalness that would normally prevent objectification, and therefore inquiry. In this article, temporal dimensions of practice are key in accomplishing this aim. In the next section, I move further into temporality in PAL and discuss some additional differences between PAL and alternate approaches to language.

A preliminary typology of classifiers in signed languages

Classifiers in spoken and signed languages are dissimilar in both form and function. However, grounds for comparison is based on the presence of morphology that classifies one of the verb's nominal arguments according to semantic criteria (Aronoff, Meir, Padden, & Sandler Reference Aronoff, Meir, Padden and Sandler2000). Many aspects of this comparison have become contentious, including characterizing classifiers, or aspects therein, as morphological, as gestural, or as some combination of the two.Footnote ¹⁰ At least three categories of signed language classifiers have been proposed (Schick Reference Schick1990; Engberg-Pedersen Reference Engberg-Pedersen1993; Aronoff et al. Reference Aronoff, Meir, Padden and Sandler2000:67; Schembri, Jones, & Burnham Reference Schembri, Jones and Burnham2005). First, there are size-and-shape specifiers, which represent shapes, outlines, or relative sizes of objects. In the following images (Figures 1 and 2), the interpreter on the right is describing nearby apartment buildings in Tactile ASL. He uses a B-handshapeFootnote ¹¹ to represent the height and shape of the building, and then to represent the sloped roof.

Figure 1.

B-handshape representing the vertical sides of the building.

Figure 2.

B-handshape representing one vertical wall and the sloped roof that meets the wall.

The second kind of classifier is an entity classifier, also known as “verbs of motion and location,” which represent categories of noun objects, often moving in a particular manner and/or direction. An example commonly cited incorporates the hand shape of the number 3,Footnote ¹² and is used to represent boats and ground vehicles (see Aronoff et al. Reference Aronoff, Meir, Padden and Sandler2000:67). In Figure 3, the interpreter is describing the relative movement of two ferries across a body of water.

Figure 3.

3-handshape representing two boats moving across a body of water.

The third kind of classifier is the handling classifier, which represents a hand manipulating an object. In Figure 4, the interpreter is describing a cooking demonstration. The cook scoops peanut butter up with a spoon, turns it over, and then tries to get the peanut butter to fall off into the bowl, but finds that it is stuck. The interpreter's hand represents the handling or manipulating of the spoon, as opposed to some aspect of the spoon itself.

Figure 4.

The interpreter “handling” a spoon.

One problem with this typology, evident at the outset, is that it conflicts with the definition that served as grounds for comparison to begin with. Formally, the comparison between spoken and signed language classifiers requires morphology, incorporated into the verb complex that classifies a nominal argument according to semantic criteria. It is not clear in these examples what constitutes the “morphology”Footnote ¹³ or the “verb complex,” nor is it clear the degree to which “classification” of the nominal argument is systematic or based on semantic criteria. Although the B-handshape contrasts systematically with other handshapes such as the C-handshape to distinguish between shapes (i.e. flat ≠ round), the path-movement and orientation of the handshape relative to the body do not vary systematically in the same manner. The meanings of the path-movement and orientation are derived as much from the shape of the building, or at least from its momentary cognitive salience, as from language-internal systematics. This leads toward a cognitive explanation, like the definition of classifiers attributed to Grinevald (Reference Grinevald1996) by Schembri (Reference Schembri2003:21) who states, “a classifier classifies in the sense that it denotes some salient inherent or perceived characteristic of the referent represented by an associated noun.” In response to analyses like these, however, Zeshan argues that cognitive classification is a pervasive phenomenon, and yet “we do not speak of classification in a linguistic sense in all cases of cognitive classification” (2003:117). To narrow cognitive classification enough to consider it a systematic linguistic phenomenon, there must be identifiable, formal objects of analysis: “no classification without classifiers” (Zeshan Reference Zeshan2003:117).Footnote ¹⁴ Since, in the case of size and shape specifiers (SASS), we cannot account for the entire form according to strictly semantic criteria, it is difficult to achieve a convincing formal analysis. Similar problems arise in all three categories of classifiers. In the next sections, each category of signed language classifier will be considered from a practice approach.

SASS classifiers in a practice approach

From the perspective of PAL, the features of SASS classifiers described above would be neither surprising nor problematic. PAL predicts that formal analysis will only ever account for part of the meanings that arise out of communicative action. The semantic field is not expected to account for more than highly schematic form-meaning correspondences in a field of contrastive oppositions. Recall that the semantic field includes “any structured set of terms that jointly subdivide a coherent space of meaning… insofar as they define a space of oppositions linked to contrasting linguistic forms” (Hanks Reference Hanks2005a:192). The semantic field of color, for example, is structured according to oppositions such as white ≠ black ≠ red (ibid.). Handshapes, as a parameter of SASS classifier constructions, when considered only within the semantic field, jointly subdivide a field of possible sizes and shapes. In the example above, two B-handshapes are used to characterize the sloped roof of a building. The sloped roof is itself made up of two flat sides. In ASL, flat ≠ round. Oppositions like this allow the interpreter to characterize the referent according to conventional categories that are available prior to the interaction. However, when the path-movement and orientation parameters of the classifier construction are considered, the form not only has a characterizing function, but also an indexical function. The first function derives from the semantic field, and the second derives from the deictic field. Analytically distinguishing between these fields means that it is not necessary account for deictic facts with semantic principles. In the deictic field, the building is individuated as an object of reference relative to the Deaf-Blind person and his interpreter. Recall that deictic reference, in Hanks' scheme, has the following structure (2009:12), given in (5) below.

In this case, the speaker and addressee have nonreciprocal sensory access to the object. The addressee has (i) his memory of what a high rise building and a sloped roof looks like; (ii) probably some outlines, or contrasts of light and dark, that he can grasp visually; and (iii) the conventional category selected by the interpreter that can be used to integrate (i) and (ii). The signer has visual access to the object, which despite the tacit character of his visual perception, probably involves similar kinds of integration. These processes cannot be accounted for with the semantics of the language. Furthermore, part of what grounds reference is the activity frame of tactile interpreting itself, which was established through sociopolitical processes beginning in the 1980s. The way in which meaning accrues to form in the deictic field cannot be understood without recognition of the sociohistorical field out of which tactile interpreting practices, and participation frameworks, emerged. This accounts for how the fairly abstract roles of speaker and addressee are transformed into the historically and socially specific roles of Deaf-Blind person and tactile interpreter, which are in turn a precondition for the availability to the Deaf-Blind person of subsequent social roles. For example, individuating the building might allow the Deaf-Blind person to inhabit the social position of a sightseer, or a dog walker, or a man in the park.

Embedding along temporal dimensions

In the following sections, PAL is extended to account for particular temporal features of the communicative events under consideration. It is argued that articulatory time, lived time, and historical time, as dimensions of the semantic, deictic, and social field, respectively, are embeddedFootnote ¹⁶ to yield an immediate surround that is accessible to both the Deaf-Blind person and the interpreter. In other words, the term “embedding” here describes the synthesis and negation of (i) the temporal regularities of sign production in Tactile American Sign Language, (ii) the temporalities of movements or actions in the deictic field, and (iii) the temporal structures which derive from the “rhythms of everyday life” or the postindustrial, urban field of rhythm. Recall that embedding in practice begins with the social field. Temporal sensibilities that accrue to the Deaf-Blind person as a modern, urban subject are a precondition for the intelligibility of particular kinds of rhythmic activity in the immediate surround (Lefebvre Reference Lefebvre2004). From the perspective of the analyst, though, the temporality of articulation in producing schematic elements of classifier constructions is posited a priori. This temporality is then embedded in representations of temporalities in the deictic field, which are further embedded in temporalities ordered by distinctions in the social field. After the schematic elements of the classifier and classifier-like constructions have been posited and viewed in their embedded state, they are retrieved and re-examined to determine the ways in which they have been transformed. This analytic process of embedding should be understood in contrast to modes of analysis that take categories such as morphology and gesture as two salient layers or parts, which are added onto one another and taken apart again, without consequence for form or meaning.

Methodology and relevance to the literature on Deaf-Blind communication

The linguistic anthropological approach taken in this research diverges from approaches taken in prior work on communication practices among Deaf-Blind people. Apart from biographical accounts of the lives of Deaf-Blind individuals such as Helen Keller (Davidson Reference Davidson1969; Herrmann Reference Herrmann1998) and Laura Bridgman (Freeburg Reference Freeburg2001; Gitter Reference Gitter2001), the bulk of research about Deaf-Blind people comes out of the fields of education and vocational rehabilitation, primarily oriented toward teaching and counseling methods for individuals (Bruce 2002; Bourquin & Sauerburger Reference Bourquin and Sauerburger2005; Janssen, Riksen-Walraven, & van Dijk Reference Janssen, Riksen-Walraven and van Dijk2003 Nelson Reference Nelson2005; Arthur-Kelley, Foreman, Bennett, & Pascoe Reference Arthur-Kelley, Foreman, Bennett and Pascoe2008; Correa-Torres Reference Correa-Torres2008; Dammeyer Reference Dammeyer2009; Parker Reference Parker2009) and practical instruction for signed language interpreters (Sauerburger Reference Sauerburger1993; Smith Reference Smith1994; Frankel Reference Frankel2002; Metzger, Fleetwood, & Collins Reference Metzger, Fleetwood; and Collins2004; Nuccio & Smith Reference Nuccio and Smith2010). This project builds, instead, on the few studies that have sought to describe language and interaction among Deaf-Blind people (Collins & Petronio Reference Collins, Petronio and Lucas1998; Mesch Reference Mesch2001; Collins Reference Collins and Petronio2004; Petronio & Dively Reference Petronio and Dively2006). However, my research differs from these studies in two respects. First, following traditional methods in linguistics and sociology, previous researchers chose participants who best approximated “ideal speaker-listeners.” This project focuses instead on the processes through which Deaf-Blind people adjust to persistent changes in sensory orientation. This includes periods of time when individuals are comfortable using language, and interacting in their habitual environments, but also times when they are struggling to identify ordinary objects and produce and receive communicative signals. Convention in the Seattle Deaf-Blind community does not demand identical sensory experience or forms of language-use among its members, but shared ways of accommodating changing and nonreciprocal sensory orientations.

All of the data directly cited in this article were collected in a four-month period of fieldwork in the spring of 2008.Footnote ¹⁷ In this study, eight dyads composed of one Deaf-Blind person and one sighted person (either deaf or hearing) were video recorded for 1.5–3.0 hours engaging in a variety of activities such as dog walking, grocery shopping, or attending an event. For those interactions where the subjects were walking, I walked in front of them and recorded them with a camera mounted on a harness and pointed backward over my shoulder. Field notes were collected throughout, and these notes form the basis for much of the ethnographic description. I also conducted interviews with Deaf-Blind people in order to understand their perspectives on how interpreters who interpret visual information are useful in building a tactile world. I selectively transcribed the video data using a transcription program called ELAN,Footnote ¹⁸ slowing the speed of interaction and glossing bodily movements, environmental details, utterances, and gestures. However, I also draw on longer-term ethnographic involvement and interviews with members of this community.

In the examples presented here, a Deaf-Blind woman, Chantelle,Footnote ¹⁹ and a tactile interpreter, Deanna, with whom Chantelle worked many times before are visiting the Olympic Sculpture Park in Seattle. In a separate interview with Chantelle, I asked if detailed information about people's bodily features, faces, and orientations in space had been useful to her when incorporated into tactile translations. A translation of her answer is given below in (6).Footnote ²⁰

1. (6) It's really useful when the interpreter comments on the environment at the same time I am looking at it. I think it's because when my eyes take in information, my brain mis-interprets it. So I have to try really hard to ignore the information my eyes are giving me, and focus on the information the interpreter is giving me through my hands. Then I can start to use my eyes as a secondary source of information, rather than a primary source, and my brain and the environment start exchanging signals again. I should have realized that sooner. It's the same thing with facial expressions—you can interpret furrowed brows in any number of ways. Maybe the person is frustrated, or sad, or emotional. But does that mean that the furrowed brows always mean the same thing? No. As your vision starts to go, and you are relying on just what details you can see, you start to realize that this kind of interpretation of cues is complicated. When I get information through my hands from an interpreter, I can cross reference that with my new vision, my memories of what I knew before with my old vision, and I can understand what is going on.

An important part of orienting to the immediate environment is building a horizon of potential. This requires the ability to link typical categories to past instances, and to anticipate potential, future instances (see Hanks Reference Hanks1996:131). What Chantelle is saying in (6) is that the inner horizon of typesFootnote ²¹ (linguistic, perceptual, and social) cannot be located with her eyes alone because her vision has changed too much. She is losing the ability to compare past tokens remembered through her past vision, with current tokens that she knows with her current vision and her tactile sensibilities. One way she addresses this problem is to work with interpreters. In these interactions, she is not merely “gaining access” to a given world, but correlating her past system of relevances with the vivid experiences of a world overfull with specificity. The analysis that follows, then, seeks to understand how Chantelle and her interpreter find relevant relations between elements in distinct fields in order to link aspects of past tokens with aspects of current tokens. In the process, I ask if the three mechanisms posited thus far by Hanks—practical equivalences, counterparts, and rules of thumb—account for the observed embeddings in interactions between Chantelle and Deanna.

Finding a way in: Integration as embedding

In the first video clip, Chantelle and Deanna are sitting at the edge of a park watching people go by on a trail that runs along the waterfront in downtown Seattle. They chose this park because they had never been there and wanted to know what kind of place it was. They also wanted to be a part of what was going on outdoors on one of the first sunny days of the year in Seattle. As they approached the park, Deanna told Chantelle that there were many couples out walking around together, smiling. Chantelle said that last week everyone had been walking around frowning, but now that the sun was out, everyone was smiling! She stopped walking and sort of hopped in place saying the sun was making her so happy. A few days after this, I went to a dog park with Helen, another Deaf-Blind woman in the community, and her regular interpreter, Kate. While Kate was looking for parking, Helen gave me an in-depth debriefing on the difference between the kind of vitamin D you get from food and the kind of vitamin D you get from the sun. She had been reading up on it because she was a biologist and wanted to know the biological explanation for why the sun had been making everyone so happy. She explained that it was a physiological response to the end of a vitamin D deficiency— the kind of vitamin D you can only get from the sun. The sun was a topic of conversation around this time, and not only for members of the Seattle Deaf-Blind community. Every year at the end of the winter in Seattle, there are diagnoses, rationalizations, and realizations. When the sun comes out, burdens become lighter, decisions are reconsidered, and people are reminded that there are true physiological effects from the absence of sunlight. I emphasize this because in Seattle, and no doubt in other places where there are long, wet winters, the first days of sun are important and are a part of many encounters that take place. We return to this point in the analysis below.

In Figure 5,Footnote ²² the interpreter (on the right) begins by signing two iterations of the sign CYCLE, indicating a particular speed, followed by a sign that is often glossed as GONE but that suggests a quick leaving (something leaving so quickly that it is already gone). This second sign traces the direction of the leaving as well as its speed. Lastly, the interpreter signs FIRST and SECOND, following the path-movements of the cyclists relative to one another with the articulation of each number.

Figure 5.

Deanna and Chantelle individuate an object of reference. (See online: http://dl.dropbox.com/u/3711257/Edwards_2012_Movie_1.mov)

The form-meaning correspondences that can be posited a priori are the signs pictured in the frames above: CYCLE, GONE, FIRST, and SECOND, as well as the handshape Chantelle uses to point with. Each one of these first four signs has a conventional meaning (which the glosses are meant to approximate).Footnote ²³ The formal aspects of the signs that are linked to those meanings include handshape, location, orientation, and movement. In order to distinguish between relations like these that inhere in the semantic field and those that inhere in the deictic field, I would like to start with some categories Liddell (Reference Liddell and McNeill2000) proposes for Visual American Sign Language.

Liddell makes a three-way distinction between categories of signs in American Sign Language based on their spatial relation to the body of the signer, and to referents that influence the location of articulation. The first category is characterized by articulatory uses of space, where a change in location results in an entirely different sign. His example is the pair ONION/APPLE. These signs are identical in all respects except the location of contact on the body of the signer. The former is articulated on the cheekbone, and the latter is articulated on the lower cheek. If the location moves to the chin, a “possible but non-existent sign” is produced (Liddell Reference Liddell and McNeill2000:332). The second category of signs, rather than coming into contact with a particular location on the body or face, is articulated in a particular location in space relative to the body. An example of this is the pair POINT/GOAL. These signs are, again, identical in all respects but location. The former is articulated at the level of the abdomen while the latter is articulated at the level of the forehead. The third category, which includes deictic gestures, “can be made in what appears to be an unlimited number of locations”Footnote ²⁴ (Liddell Reference Liddell and McNeill2000:333). In contrast to the first two categories,

Liddell does not include temporality, which is incorporated into the “movement” rather than the “location” parameter of the sign. However, temporal dimensions of the first two categories of signs differ from that of the third, just as the spatial aspects do. In the first categories, the temporality of articulation derives from language-internal distinctions. In the third case, the temporality of signs that are directed toward or motivated by movements of referents in the immediate environment derive from relations in the deictic field.

In Liddell's typology, the signs used by Deanna would normally be excluded from the third category. The citation form of CYCLE is conventionally articulated in front of the chest without contact with the body. The citation form of GONE is conventionally articulated in a path beginning near the face and moving forward. The citation forms of FIRST and SECOND are conventionally articulated in neutral signing space in front of the torso. None of these signs require a deictic construal to be meaningful. Once they are embedded in the deictic field, however, the spatiotemporal properties of the location and movement parameters shift, taking on properties of signs in Liddell's third category. The temporal features of CYCLE that inhere in the semantic field include articulatory time, or the temporal aspect of the movement required for production, and an aspectual distinction, which signals that the represented event is ongoing. This is a documented aspectual function in at least some signed languages (Rathmann Reference Rathmann2005). Both of these temporal features characterize the observed event according to language-internal distinctions. In addition, there is a temporal dimension that indexes the speed and rhythm of the cyclists.

As Deanna is signing GONE GONE, Chantelle points toward an object, which Deanna takes up as a question about the location of the referent. Deanna confirms that the direction Chantelle is pointing in is in the direction of the referent by aligning the handshape of the next sign FIRST with Chantelle's outstretched arm. The pointing handshape functions both as confirmation of Chantelle's localization of the referent and the beginning of the sign FIRST. SECOND is then articulated in a spatial relation to FIRST such that the path between them aligns with the path-movement of GONE GONE. FIRST is articulated at the end of the path established by GONE, and SECOND is articulated at the beginning (the first cyclist is further along the path than the second). As described in the interview segment above, Chantelle is triangulating here between (i) visual memory of what cyclists looks like, (ii) the degraded visual image of the cyclists she and Deanna are pointing to, and (iii) the tactile input given by Deanna.

However, there is more to the individuation of an object of reference than its naming or its localization in space. Part of what grounds reference, at least for the interpreter as a sighted person, is that the cyclists are typified in more fine-grained terms than the category “cyclist” entails. It is not uncommon to see cyclists riding by in a public place in Seattle, and out of that regularity, types of cyclists emerge. These cyclists are of the serious variety—they are agile, riding in time with one another, moving at a good clip through a crowded area. They are attired in spandex, cycling shoes, and aerodynamic sunglasses—they aren't wearing jeans or carrying a box of groceries balanced on their back fenders. For Chantelle, it is precisely this link between instances and the typical categories to which they refer that is disrupted. How, then, do Chantelle and Deanna, who have nonreciprocal access to the deictic field, establish a horizon against which object can be individuated? What are the relations between semantic, deictic, and social fields that yield a common structure of relevance?

My claim is that because temporality is as much a feature of tactile worlds as it is of visual worlds, it offers a dimension that articulates to fields organized along tactile lines, as well as those organized along visual lines. For Chantelle, motion is no longer a visual phenomenon, but her feet still hit the pavement at regular intervals when she walks; tactile language is still ordered according to particular patterns of motion and rest; the work week still trudges on until the weekend breaks it up; and the cloistered months of rain still open onto a world of possibility on the first sunny days of the year. These regularities converge in unrepeatable, accidental unities that create a sense of being somewhere in particular, as opposed to merely being in an example of a place, or a type of place.

In the portion of the interaction examined thus far, regularities derive from fields that can be analytically distinguished. Patterns of motion and rest are a feature of articulation, which is part of the ground against which other temporal regularities cohere. Temporal features of the deictic field are also characterized explicitly. In American Sign Language GONE ≠ GO ≠ LEAVE. This set of terms jointly subdivides a space of meaning that can be described as “speed of departure.” There is a temporality and an unmarked path-movement associated with the citation forms of these signs. In all cases, the path-movement begins near the body and moves forward into neutral signing space. Once GONE is embedded in the deictic field, its path-movement shifts. Rather than beginning near the face and moving away into the signing space, it is produced horizontally, indicating an idealized path of the cyclists who have just gone by. At this point, the relation has been established in the articulation of the sign CYCLE, the sign GONE, and the pair FIRST, SECOND. In each case, there are shifts in temporal regularity once the sign is embedded in the deictic field. Rather than being articulated in neutral signing space, FIRST and SECOND are produced in two locations along a path that corresponds with the paths of the cyclists and the path of articulation of the signs GONE, GONE. Lastly, all three pairs of signs sync rhythmically with one another: CYCLE CYCLE, GONE GONE, FIRST SECOND—1, 2; 1, 2; 1, 2.

These rhythmic patterns in language can articulate to fields organized along visual lines just as they can articulate to fields organized along tactile lines. Having lived in a visual world most of her life, every time Chantelle remembers or refers to a visual scene, the action of articulation in referring has both a visual dimension (in memory or sight) and a haptic dimension (in the expression of that memory or sight in American Sign Language). In this way, American Sign Language, whether instantiated in a visual world or a tactile world, has a tactile dimension for sighted and blind people alike. In Chantelle's case, this formal continuity is still part of her corporeal schema, and it comes into play in efforts to translate across visual and tactile modes of knowing. The integration of articulatory time with the felt time of the cyclists in the deictic field constitutes a relation between the semantic and deictic fields that is accessible to both Chantelle and Deanna. The challenge from there lies not in localizing or describing elements in the deictic field, but in linking semantico-deictic structures, established in interaction, to the social field. As the interaction unfolds further, we see how rhythmic repetition further seals the semantic and deictic fields to one another by producing a kind of density that accrues to the relations between these fields. Further, we see how this density leads Chantelle into a problem space where she can inquire about relations between semantico-deictic structures and the social field, if not establish them outright.

In the second clip (Figure 6),Footnote ²⁵ Deanna points out a nearby woman walking by and incorporates temporal dimensions of her speed and direction, explaining that the woman is also talking on a phone. There are three iterations of WALK, which jointly signify that the represented event is ongoing. The interpreter points both before and after the sign WALK, tracing a path corresponding to the path-movement of the walker. The spatial and temporal features of WALK, which derive from distinctions in the semantic field are transformed upon embedding in the deictic field. Because WALK is not phonologically amenable to internal spatial modifications the way it is amenable to temporal modifications, the pointing action occurs sequentially.

Figure 6.

Mutually available deictic relations begin to diverge for Chantelle and Deanna in the social field. (See online: http://dl.dropbox.com/u/3711257/Edwards_2012_Movie_2_.mov)

Again, the citation form of WALK is not treated as a deictic. As it is instantiated here, however, it is transformed as it is embedded in the deictic field to index spatiotemporal dimensions of the immediate surround. Therefore WALK moves into Liddell's third category of signs, in which “the direction of the movement of the sign [is] different, depending on the location of the person toward whom the sign [is] directed” (2000:333).

If we consider the broader field of postindustrial urban rhythm, interpretive possibilities are further restricted. This is the rhythm of fe2et that know where they are going, attached to a person with a purpose in mind. They are the feet of a local, maybe an athlete, not a Sunday dog walker, a window shopper, or a tourist. These are not the rhythms of articulation, or the patterns of motion and rest in the deictic field. Modern time, according to Lefebvre (Reference Lefebvre2004:74), “divides itself into lots and parcels: … various forms of work, entertainment, and leisure… [E]very ‘doing’ has its time.” These temporal lots and parcels converge in space in semistructured but ultimately unpredictable accidents, forming what Lefebvre calls, the “rhythms of everyday life.” The integration of these rhythms into the patterns of motion and rest in the language and in the deictic field can literally be felt in the reception of the translation. Insofar as these rhythms articulate to both visual and tactile fields, they can serve as points of entry for both parties to the interaction.

There is a sequential, temporal organization in the back-and-forth of conversation between Chantelle and Deanna that is consequential as well—a temporal order Enfield calls “enchronic” (Reference Enfield2009:10). Deanna says there is a woman who is walking (which she has said of other people) and then says BUT, followed by TALK-ON-PHONE, TALK-ON-CELL and TALK. BUT indicates that what will follow is a unique characteristic of the referent, something that distinguishes her from the other people walking in the park. TALK-ON-PHONE is signed three times, using two different handshapes (glossed as both TALK-ON-PHONE and TALK-ON-CELL-PHONE). Each time she signs TALK-ON-PHONE, Deanna pauses with the phone by the ear. After she signs TALK, she pauses again. These pauses are taken up by Chantelle as a request for commentary, and she responds by saying REALLY and mimicking a woman talking on the phone. Then her mouth drops open and she leans toward Deanna.

In the third clip (Figure 7),Footnote ²⁶ Deanna repeats what she has just said, with a few changes. She signs WALK slightly faster, and bookends the description with evaluative statements: at the beginning with an expression that suggests reflection, roughly equivalent to saying “hmmmm” or “wow”Footnote ²⁷ and at the end, with the sign FUNNY (adding up to something like “wow, that woman is out to get some exercise, and meanwhile is talking away on her cell phone. That's funny.”). Although Deanna does not sign “exercise,” this is a possible interpretation of the rhythm of articulation in WALK. In the second utterance, Deanna signs WALK with more strength and speed. Chantelle again has been put in a position to comment, so she says, “Really? New technology! My god!” Whatever Deanna expected Chantelle to say, this was not it, as the indeterminate gesture Deanna makes in the next turn reveals. Chantelle's response has disrupted the steady back-and-forth of conversation, which has in turn, opened up an inquiry into the relation between the semantico-deictic structures established thus far and the social field. More specifically, this brings them from articulatory time, used to represent spatiotemporal features of the immediate surround, into the social field of technological innovation, which is intimately linked to the temporal orders already established. Chantelle interprets Deanna's gesture as meaning that the technology is not in fact new and that perhaps Chantelle should have known this. Chantelle says, “Wow, I'm behind on technology.”

Figure 7.

Individuation of object allows Chantelle to inquire into relations between deictic and social fields. (See online: http://dl.dropbox.com/u/3711257/Edwards_2012_Movie_3.mov)

The temporality of technological innovation, and the way it is recognized in space, is a feature of the social field. Failing to keep up, as is clear in this example, can engender what Lefebvre would call an arrhythmic response (2004:29): a moment of shock in which the ongoing synthesis of linguistic, physical, and social fragments is halted as when Chantelle reflects that she has “fallen behind on technology.” This realization is not exclusively discursive. It affects her embodied orientation to the immediate environment, and as such is highly consequential for her orientation to the deictic field.Footnote ²⁸ The possibility of exercising while talking on the phone integrated into Chantelle's structures of anticipation in her everyday life prompts a broader restructuring of an anticipatory field encapsulating the Olympic Sculpture Park in Seattle, or park life in the city, or city life more generally. With repetition, reference to objects that can be referred to this category (a person walking in public while talking on a cell phone) will dissolve into the horizon of everyday life in the city, and Chantelle will move differently through the city. Without repetition, they will not.

In addition to the rhythms of technological innovation, the participant framework within which the interaction occurs is also part of the social field. Up until this point, Chantelle and Deanna have been occupying the roles of “Deaf-Blind person” and “interpreter,” which are given by the tactile interpreting frame, and which authorize the communicative activity of both participants. This frame does not exist everywhere that there are individuals who cannot hear or see. When Deaf-Blind people from Seattle go other places where Deaf-Blind individuals are, like meetings of the American Association of the Deaf-Blind (AADB), communication is frustrating and slow, and the degree to which Deaf-Blind conversation and interaction are not conventionalized outside of Seattle becomes apparent. Communicative practices have become conventional as a result of social and political action and have been subject to historical contingencies. They are not merely the interactional outcome of a physiological condition. Thus, these practices are shaped not only in the course of interaction, but also by prescriptive discourses about how to be a tactile interpreter, which in turn are shaped by broader legal discourses that make interpreting services like these a right.

The language of the Americans with Disabilities Act (ADA), for example, focuses on the rights of the individual, and requires that “reasonable accommodations” be made to ensure those rights. From this perspective, the Deaf-Blind person is not constituted in interaction, or by history, but is always already an individual with rights, and the interpreter is the means by which accommodation is accomplished. This discursive frame implies that the role of the interpreter is to “provide access” to a given world, and the Deaf-Blind person is merely a receptacle for “information” about that world. The ADA thus valorizes visual forms of common sense over tactile common sense. Though it is not determinant, this model shapes the interaction between Chantelle and Deanna. In referring to the woman talking on the phone, many things go unsaid by Deanna, including the common sense that a woman who exercises while talking on her cell phone is one type of woman—perhaps in this case a type that should be distinguished from us. But based on Chantelle's response, we know that this is the common sense of a visual world. Visual common sense will always, in the interpreting frame, trump the common sense of a tactile world insofar as human rights discourses such as those embedded in the ADA structure these interactions by treating knowledge acquired tactually as subordinate to knowledge acquired visually. Of course, addressing this asymmetry is not merely a matter of correcting the language of the ADA, debunking the notion of a given world, or implementing an egalitarian ideology of sensory orientation. The problem, as practice theory would predict, runs much deeper than that.

Several years ago, I was walking with a Deaf-Blind woman into an apartment building. She said she was impressed that the building was fancy. I disagreed, informing her that it wasn't fancy at all. She sighed, and instructed me to close my eyes, take her cane, and walk. As I swept the cane back and forth, slowly shuffling my feet forward, the marble tile felt immaculately clean, smooth. The cane seemed to float across it. A faint smell of perfume drifted out of the manager's office and mixed with the fading, subtle smell of soap on the floor. I conceded. And yet, as I opened my eyes to the peeling green paint on the walls and the rusting mailbox numbers, I was struck by the concreteness of the visual world I inhabit. It was difficult for me to adopt a tactile perspective, despite the evidence that it was just as coherent as mine, precisely because of the sense that what I saw was utterly concrete. In closing my eyes and feeling the texture of the tile floor with the cane, I did not develop a tactile habitus. I gained a momentary insight, which would not necessarily accrue to the horizon against which categories and relations of relevance would emerge for me in the future. The same is true to some degree for Chantelle. A momentary insight into a visual world is likely to be useful, but it is not the same thing as having “access” to it. The aim, as Chantelle stated in her interview, is to use these insights in building continuity between the fading visual world she is relinquishing and the tactile world into which she is venturing. This aim is disjunctive with the access model of disability discourse, but these discourses must be continually reproduced in order to secure funding to pay for interpreters and other services. These discourses are part of the social field rather than the semantic or deictic fields, and yet they have a direct impact on how the actual structure of the interaction unfolds and the degree to which it can serve the purposes of Deaf-Blind people in bringing a tactile world into being.

I have shown in these sections that the analytics of semantic, deictic, and social fields are all necessary to understand how Chantelle uses interpreted interactions to build an inhabitable tactile world. I have also shown how disparate elements are drawn into relation with one another through processes of embedding that rely on the integration of temporal regularities across fields. In the next section, integration is proposed as a fourth mechanism of embedding alongside counterparts, equivalences, and rules of thumb in the practice approach to language.

Integration as principle of embedding

Recall that embedding is the means by which highly schematic form-meaning mappings undergo reshaping, conversion, and transformation according to patterns that crystallize in the relations between semantic, deictic, and social fields (Hanks 2005:194). In each example above, relations between elements in distinct fields are established by embedding language-internal temporal regularities in the lived temporalities of deictic and social fields. These rhythmic composites are expressed in the movement parameter of the signs. Each composite is the outcome of a transformation of a priori categories with which the analysis began. In the instance of CYCLE, we begin in the semantic field, where articulatory time combines with repetition to yield an aspectual meaning derived from language-internal distinctions. Once CYCLE is considered in its relation to the other signs in the utterance and in its relation to the deictic field, its meaning is transformed. The signs CYCLE CYCLE, GONE GONE, FIRST SECOND must be viewed as part of a larger rhythmic pattern, which in addition to characterizing the cyclists, indexes the temporal coordination of the cyclists with one another, and their speed and their path relative to Chantelle and Deanna. There are further meaning effects when one considers the temporal regularities of CYCLE from the perspective of the social field. These are cyclists of the serious variety, agile enough to ride quickly and in sync with one another through a crowded park. Their time is the time of training or preparation more than leisure or transport. These are distinctions that cannot be found in either the semantic or the deictic fields.

The relations of embedding I have just described can be distinguished from those laid out by Hanks (Reference Hanks2005a) since, in this case, not only are the meanings converted, but also the forms. The articulation of CYCLE integrates temporal distinctions that do not derive from the language but from the fields to which linguistic distinctions articulate. It is the movement itself that is meaningful in distinct ways across three fields, and as the movement parameter is successively embedded, its form is reshaped along with its meaning. Because of this formal reshaping, neither a practical equivalence nor a counterpart can account for the relation that is established between the fields. This is because in each of these relations some formal element must remain constant across contexts. To capture this difference, I propose a relation of integration, whereby formal aspects in each of the embedded fields are mutually transformed along with their meanings.

Counterparts are identity relations between objects, which are established in referring to them, or construing them using the same linguistic form (Hanks Reference Hanks2005a:202). Therefore, a counterpart relation requires a stable, mediating linguistic unit. Likewise, practical equivalences establish correspondences between modes of access according to a stable, mediating formal unit. In these examples, temporal elements in distinct fields are integrated into a unified temporality, one that is not equivalent in either form or meaning to the elements that gave rise to it. Lastly, rules of thumb emerge out of a kind of regularity that is not a characteristic of Chantelle's ordinary interactions at this point in her transition to tactility. The naturalness of encounters, such as greetings and their conventional linkage to certain phrases, is precisely what is at stake in the work she is doing to build a world. Therefore, rules of thumb do not account for the links between fields that emerge out of this communicative context. In contrast to counterparts, practical equivalences, and rules of thumb, integration implies a partial projection of elements from two domains into a third. The third domain manifests a structure that is not present in either of its inputs, and in this sense is emergent (Fauconnier & Turner Reference Fauconnier and Turner1998:133; Fauconnier & Turner Reference Fauconnier and Turner2002).Footnote ²⁹ In the examples described here, formal and conceptual elements of each sign are selectively incorporated into the formal and conceptual structure of the embedded sign, thereby manifesting a relation of integration.